In this study we characterized children exposure to extremely low frequency (ELF) magnetic fields using cluster analysis - a Machine Learning approach. Indoor personal exposure measurements from 977 children in France were analyzed to discover how electric networks near child home or school could influence exposure patterns. 225 kV/400 kV overhead lines characterized the cluster of children with the highest exposure; 63 kV/150 kV overhead lines characterized the cluster with mid-to-high exposure; 400 V/20 kV substations and underground networks characterized mid-to-low exposures. 400 V/20 kV overhead lines and 63-225 kV underground networks had a marginal contribution in differentiating and characterizing the exposure clusters.
Unsupervised Machine Learning techniques for the characterization of children exposure to ELF MF
Gabriella Tognola;Marta Bonato;Emma Chiaramello;Serena Fiocchi;Marta Parazzini;
2019
Abstract
In this study we characterized children exposure to extremely low frequency (ELF) magnetic fields using cluster analysis - a Machine Learning approach. Indoor personal exposure measurements from 977 children in France were analyzed to discover how electric networks near child home or school could influence exposure patterns. 225 kV/400 kV overhead lines characterized the cluster of children with the highest exposure; 63 kV/150 kV overhead lines characterized the cluster with mid-to-high exposure; 400 V/20 kV substations and underground networks characterized mid-to-low exposures. 400 V/20 kV overhead lines and 63-225 kV underground networks had a marginal contribution in differentiating and characterizing the exposure clusters.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
